3. Data, Science & AI

Manages vector embeddings for semantic search in CarInsight, enabling efficient text retrieval and matching capabilities.

Provides practical guidance for building and maintaining production-ready machine learning systems, covering evaluation, deployment, monitoring, and MLOps best practices.

Analyzes agent session logs, progress.md, and git history to identify patterns and provide actionable insights for improving future agent performance.

Creates AI agents with integrated monitoring for observability, KPI tracking, and governance in AI workflows.

Generates analytics reports from Olakai data via CLI, providing usage summaries, KPI trends, risk analysis, and efficiency insights without web UI access.

Manages AI prompt configurations, versions, and model bindings for the KR92 Bible Voice AI system across multiple vendors and models.

Enables persistent memory for AI agents, recalling historical context at session start and permanently recording critical decisions.

Validates TV show data (shows, characters, episodes, quotes) against FanHub's domain rules to ensure data integrity and business rule compliance.

Designs and optimizes database schemas, migrations, and queries for relational and document databases.

Enables data analysis, scientific computing, and visualization using pandas, numpy, and matplotlib.

Provides Bible reference lookups, validation, and OSIS format handling for the KR92 Bible Voice project.

Provides session history retrieval capabilities for AI agents, designed to be invoked via the hive-mind:retrieval agent for efficient information access.

Executes Stata code via temporary do-file, ensuring correct graph rendering and multi-line command handling for statistical analysis.

Converts analysis reports from documentation into structured tasks stored in a SQLite database for task management.

Performs data processing, analysis, and visualization using Python data science libraries for actionable insights.

Opens the tasks database using DB Browser for SQLite for efficient data querying and management.

Designs structured MCP resources to expose content for efficient LLM consumption in static or dynamic xmcp configurations.

Integrates Google Gemini Flash API for automated data extraction from passport and license documents using OCR.

Designs reusable parameterized prompt templates for MCP in xmcp, optimizing AI model interaction efficiency.

Integrates Google Gemini Flash API for document OCR, extracting structured data from passports and licenses.

Links entity triplets to central knowledge nodes, enabling integration of multiple knowledge layers for enhanced data relationships.

Converts verbose memory data into SSL v0.2 format, functioning as both encoder and decoder for structured data representation.

Runs the LoCoMo benchmark to assess long-term conversational memory performance in AI models.

Provides comprehensive tools for creating, editing, and analyzing spreadsheets with formulas, formatting, and data visualization across multiple formats.